• State Resolved
  • Locked Locked
  • Replies 4 replies
  • Subscribers 63 subscribers
  • Views 298 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

BQ76952: OCD1 Delay setting worsens OCD1 Threshold value

Andras Magyar
Intellectual 400 points
Part Number: BQ76952

Dear Experts,

I observed that when I set OCD1 delay between 20-max ms the accuracy of the OCD1 Threshold worsens with a few Ampers plus above the threshold.

Is it something expected?

Thanks and Regards,

Andras

  • 0
    TI__Expert 5090 points

    Hi Andras,

    We do not change the threshold directly with a change in  the delay setting.  Have you noticed if the OCD1 Safety Alert threshold changes as well, or does this only affect when the OCD1 Safety Status is triggered?

    To be clear, what delays have you tested?   I'm thinking this might be an artifact of the statistical distribution of the noise in the system.  When you have the delay set to 20ms, it only requires ~6 consecutive checks overthreshold to trigger the safety status.  When you have the delay set much longer, it requires many more consecutive checks overthreshold, which gives it many more opportunities to have one check underthreshold and clear the timer.  So then it may result in you needing an input farther overthreshold to avoid the noise clearing the timer before it completes.

    Thanks,

    Terry

  • 0 in reply to -Terry
    Intellectual 400 points

    Hi Terry,

    It is very interesting because now I tested two scenarios:

    When I set the delay to 330ms the Safety Alert A (OCD1 bit) starts to go on and off around the threshold and it takes 1-1.5A current above the threshold to get stabilized but even if the alert is hit constantly the status is not set after 330ms because it takes another 1A more or less to hit the status permanently and the fets to go off (As I said earlier (threshold+few amps).

    When I set the delay to 10ms the Safety Alert A (OCD1 bit) is on and off a bit below the threshold but one the threshold is hit Safety Status A OCD1 bit gets stabilized and fets go off.

    My setup is that I have a shunt resistor with 0.3mOhm value sleep mode off.

    By the way I experience the same thing with OCC protection.

    Thanks,

    Andras

  • +1 in reply to Andras Magyar
    TI__Expert 5090 points

    Hi Andras,

    It sounds like the noise issue I mentioned above.  Since the device evaluates the overcurrent case every ~3.3ms, all it takes is one evaluation below threshold to reset the delay (I'm not sure if you are really reading the safety alert fast enough to capture every 3.3ms evaluation result). 

    When you have a very long delay (330ms), it needs around 100 evaluations in a row to be over threshold before the safety status will trigger.  But when you have a short delay (10ms), you only need 3 evaluations in a row to be over threshold to get the safety status. 

    You may need to adjust either the threshold or the delay to account for this - you can use the shorter delay to have the threshold be closer to what is expected, or if you really want to use a longer delay, then you can adjust the threshold for your desired level, based on your own experiments.

    Thanks,

    Terry

  • 0 in reply to -Terry
    Intellectual 400 points

    Hi Terry,

    Yes, I'll set the registers based on my validations.

    Thanks and Regards,

    Andras